You are here:

An investigation of segmentation analysis modeling as a knowledge discovery technique for the planning and the design of instructional technology research DISSERTATION

, Wayne State University, United States

Wayne State University . Awarded

Abstract

This study introduces the concept of computer-assisted segmentation analysis modeling (SAM) as a knowledge discovery technique for planning of research in the social sciences, particularly in the field of instructional technology. This investigative study of SAM develops each step in the process, considers major methodological issues and problems, and evaluates computer programs which can be used to facilitate the analysis.

Purposively selected examples from seven doctoral studies of instructional technology at a major Midwestern university are analyzed using an ex-post facto, multiple case contrast design. The SAM technique uses the chi square automatic interaction detection software $\rm (AnswerTree\sp\circler )$ by SPSS, Inc. $\rm AnswerTree\sp\circler$ algorithms are a form of automatic interaction detection analysis which use a Bonferroni adjusted chi square to split samples on significant characteristics. Hierarchical classification (decision) tree diagrams are presented to map the output of segmentation analysis. These maps provide a visual framework to guide the planning and design of proposed studies. The maps are also used for reporting quantitative analysis and displaying output details.

Findings of the seven doctoral studies using linear quantitative statistical methods are revisited and then contrasted to the results of SAM on the same datasets. The exhaustive search methodology of SAM confirms the research findings and discovers additional significant relationships between characteristic segments of the data not revealed in the original studies. These relationships are potential for design and planning of new confirmatory research studies. The user-interactive SAM showed a significant reduction in time and pages of output required to identify the relationships in the data sets compared with the original study methods.

Contribution of this study is the identification and validation of a powerful knowledge discovery technique which can be applied to the planning for research studies in the social sciences, particularly, instructional technology. Researchers and graduate students in the social sciences can have access to a potent exploratory data analysis technique for rapid initial planning and design of proposed confirmatory research studies.

Citation

Barton, C.D. An investigation of segmentation analysis modeling as a knowledge discovery technique for the planning and the design of instructional technology research. Ph.D. thesis, Wayne State University. Retrieved November 17, 2018 from .

This record was imported from ProQuest on October 23, 2013. [Original Record]

Citation reproduced with permission of ProQuest LLC.

For copies of dissertations and theses: (800) 521-0600/(734) 761-4700 or https://dissexpress.umi.com

Keywords